The rapid advancements in Large Language Models (LLMs) and Large Visual-Language Models (LVLMs) have opened up new opportunities for integrating visual and linguistic modalities. However, effectively aligning these modalities remains challenging, often leading to hallucinations--where generated outputs are not grounded in the visual input--and raising safety concerns across various domains. Existing alignment methods, such as instruction tuning and preference tuning, often rely on external datasets, human annotations, or complex post-processing, which limit scalability and increase costs. To address these challenges, we propose a novel approach that generates the debiased self-judgment score, a self-evaluation metric created internally by the model without relying on external resources. This enables the model to autonomously improve alignment. Our method enhances both decoding strategies and preference tuning processes, resulting in reduced hallucinations, enhanced safety, and improved overall capability. Empirical results show that our approach significantly outperforms traditional methods, offering a more effective solution for aligning LVLMs.

The binary exponential backoff scheme is widely used in WiFi 7 and still incurs poor throughput performance under dynamic channel environments. Recent model-based approaches (e.g., non-persistent and $p$-persistent CSMA) simply optimize backoff strategies under a known and fixed node density, still leading to a large throughput loss due to inaccurate node density estimation. This paper is the first to propose LLM transformer-based in-context learning (ICL) theory for optimizing channel access. We design a transformer-based ICL optimizer to pre-collect collision-threshold data examples and a query collision case. They are constructed as a prompt as the input for the transformer to learn the pattern, which then generates a predicted contention window threshold (CWT). To train the transformer for effective ICL, we develop an efficient algorithm and guarantee a near-optimal CWT prediction within limited training steps. As it may be hard to gather perfect data examples for ICL in practice, we further extend to allow erroneous data input in the prompt. We prove that our optimizer maintains minimal prediction and throughput deviations from the optimal values. Experimental results on NS-3 further demonstrate our approach's fast convergence and near-optimal throughput over existing model-based and DRL-based approaches under unknown node densities.

--Wireless Body Area Networks (WBANs) enable continuous monitoring of physiological signals for applications ranging from chronic disease management to emergency response. Recent advances in 6G communications, post-quantum cryptography, and energy harvesting have the potential to enhance WBAN performance. However, integrating these technologies into a unified, adaptive system remains a challenge. We propose a novel Large Language Model-driven adaptive WBAN framework in which a Large Language Model acts as a cognitive control plane, coordinating routing, physical layer selection, micro-energy harvesting, and post-quantum security in real time. Our review highlights the limitations of current heuristic-based designs and outlines a research agenda for resource-constrained, 6G-ready medical systems. This approach aims to enable ultra-reliable, secure, and self-optimizing WBANs for next-generation mobile health applications.

Precise, correct feedback is crucial for effectively training large language models (LLMs) in code reinforcement learning. However, synthesizing high-quality test cases remains a profoundly challenging and unsolved problem. In this work, we present Klear-CodeTest, a comprehensive test case synthesis framework featuring rigorous verification to ensure quality and reliability of test cases. Our approach achieves broad coverage of programming problems via a novel Generator-Validation (G-V) framework, ensuring correctness through a consistency validation mechanism that verifies outputs against gold solutions. The proposed G-V framework generates comprehensive test cases including both regular and corner cases, enhancing test coverage and discriminative power for solution correctness assessment in code reinforcement learning. In addition, we design a multi-layered security sandbox system optimized for online verification platforms, guaranteeing safe and reliable code execution. Through comprehensive experiments, we demonstrate the effectiveness of our curated dataset, showing significant improvements in model performance and training stability. The source codes, curated dataset and sandbox system are available at: https://github.com/Kwai-Klear/CodeTest.

Recommender systems are indispensable for helping users navigate the immense item catalogs of modern online platforms. Recently, generative recommendation has emerged as a promising paradigm, unifying the conventional retrieve-and-rank pipeline into an end-to-end model capable of dynamic generation. However, existing generative methods are fundamentally constrained by their unsupervised tokenization, which generates semantic IDs suffering from two critical flaws: (1) they are semantically flat and uninterpretable, lacking a coherent hierarchy, and (2) they are prone to representation entanglement (i.e., ``ID collisions''), which harms recommendation accuracy and diversity. To overcome these limitations, we propose HiD-VAE, a novel framework that learns hierarchically disentangled item representations through two core innovations. First, HiD-VAE pioneers a hierarchically-supervised quantization process that aligns discrete codes with multi-level item tags, yielding more uniform and disentangled IDs. Crucially, the trained codebooks can predict hierarchical tags, providing a traceable and interpretable semantic path for each recommendation. Second, to combat representation entanglement, HiD-VAE incorporates a novel uniqueness loss that directly penalizes latent space overlap. This mechanism not only resolves the critical ID collision problem but also promotes recommendation diversity by ensuring a more comprehensive utilization of the item representation space. These high-quality, disentangled IDs provide a powerful foundation for downstream generative models. Extensive experiments on three public benchmarks validate HiD-VAE's superior performance against state-of-the-art methods. The code is available at https://anonymous.4open.science/r/HiD-VAE-84B2.

Large Reasoning Models (LRMs) have achieved remarkable success, yet they often suffer from producing unnecessary and verbose reasoning chains. We identify a core aspect of this issue as "invalid thinking" -- models tend to repeatedly double-check their work after having derived the correct answer. To address this specific inefficiency, we move beyond the general principles of Efficacy and Efficiency to propose two new, fine-grained principles: Brevity, which advocates for eliminating redundancy, and Sufficiency, which ensures critical reasoning steps are preserved. Guided by these principles, we introduce LC-R1, a post-training method based on Group Relative Policy Optimization (GRPO). LC-R1 employs a novel combination of a Length Reward for overall conciseness and a Compress Reward that is specifically designed to remove the invalid portion of the thinking process. Extensive experiments on multiple reasoning benchmarks demonstrate that LC-R1 achieves a significant reduction in sequence length (~50%) with only a marginal (~2%) drop in accuracy, achieving a favorable trade-off point on the Pareto frontier that prioritizes high compression. Our analysis further validates the robustness of LC-R1 and provides valuable insights for developing more powerful yet computationally efficient LRMs. Our code is released at https://github.com/zxiangx/LC-R1.

Automated speaking assessment (ASA) on opinion expressions is often hampered by the scarcity of labeled recordings, which restricts prompt diversity and undermines scoring reliability. To address this challenge, we propose a novel training paradigm that leverages a large language models (LLM) to generate diverse responses of a given proficiency level, converts responses into synthesized speech via speaker-aware text-to-speech synthesis, and employs a dynamic importance loss to adaptively reweight training instances based on feature distribution differences between synthesized and real speech. Subsequently, a multimodal large language model integrates aligned textual features with speech signals to predict proficiency scores directly. Experiments conducted on the L TTC dataset show that our approach outperforms methods relying on real data or conventional augmentation, effectively mitigating low-resource constraints and enabling ASA on opinion expressions with cross-modal information. Index T erms: Automated speaking assessment, Opinion Expression, Data Augmentation 1. Introduction In recent years, technologies for computer-assisted language learning (CALL), such as automated speaking assessment (ASA), have made significant strides to meet the growing demand for scalable and objective evaluation of second-language (L2) speaking proficiency in both academic and professional contexts [1, 2, 3].

Exploration is essential for general-purpose robotic learning, especially in open-ended environments where dense rewards, explicit goals, or task-specific supervision are scarce. Vision-language models (VLMs), with their semantic reasoning over objects, spatial relations, and potential outcomes, present a compelling foundation for generating high-level exploratory behaviors. However, their outputs are often ungrounded, making it difficult to determine whether imagined transitions are physically feasible or informative. To bridge the gap between imagination and execution, we present IVE (Imagine, Verify, Execute), an agentic exploration framework inspired by human curiosity. Human exploration is often driven by the desire to discover novel scene configurations and to deepen understanding of the environment. Similarly, IVE leverages VLMs to abstract RGB-D observations into semantic scene graphs, imagine novel scenes, predict their physical plausibility, and generate executable skill sequences through action tools. We evaluate IVE in both simulated and real-world tabletop environments. The results show that IVE enables more diverse and meaningful exploration than RL baselines, as evidenced by a 4.1 to 7.8x increase in the entropy of visited states. Moreover, the collected experience supports downstream learning, producing policies that closely match or exceed the performance of those trained on human-collected demonstrations.

Personal use of this material is permitted. Abstract--Human drivers possess spatial and causal intelligence, enabling them to perceive driving scenarios, anticipate hazards, and react to dynamic environments. In contrast, autonomous vehicles lack these abilities, making it challenging to manage perception-related Safety of the Intended Functionality (SOTIF) risks, especially under complex or unpredictable driving conditions. T o address this gap, we propose fine-tuning multimodal large language models (MLLMs) on a customized dataset specifically designed to capture perception-related SOTIF scenarios. Benchmarking results show that fine-tuned MLLMs achieve an 11.8% improvement in close-ended VQA accuracy and a 12.0% increase in open-ended VQA scores compared to baseline models, while maintaining real-time performance with a 0.59-second average inference time per image. We validate our approach through real-world case studies in Canada and China, where fine-tuned models correctly identify safety risks that challenge even experienced human drivers. This work represents the first application of domain-specific MLLM fine-tuning for the SOTIF domain in autonomous driving. N autonomous driving (AD), safety is commonly classified into functional safety and Safety of the Intended Functionality (SOTIF). Functional safety concerns failures in hardware or software that result in unsafe operation. In contrast, SOTIF addresses hazards that occur not due to malfunctions, but when the system operates as intended yet produces unsafe outcomes because of external factors or inherent limitations [1]. Perception systems in autonomous vehicles (A Vs), which are tasked with detecting, classifying, and predicting based on environmental stimuli, are particularly vulnerable to SOTIF-related challenges. Manuscript received 2 February, 2025; revised 27 August, 2025; accepted 7 September, 2025. Y ang, and A. Khajepour are with MVS-Lab, Department of Mechanical and Mechatronics Engineering, University of Waterloo, 200 University Ave West, Waterloo ON, N2L3G1 Canada. S. Huang, and F. Shi are with CompLING Lab, David R. Cheriton School of Computer Science, University of Waterloo, 200 University Ave West, Waterloo ON, N2L3G1 Canada and V ector Institute, Toronto, Canada C. Sun is with the Department of Data and Systems Engineering, University of Hong Kong, Pok Fu Lam, Hong Kong, China (e-mail: c87sun@hku.hk) Lu is with the Department of Mechanical Engineering, University of New Brunswick, Fredericton, NB E3B 5A3, Canada (e-mail: yukun.lu@unb.ca) H. Wang is with School of V ehicle and Mobility, Tsinghua University, Beijing, China, 100084.

LLM-integrated app systems extend the utility of Large Language Models (LLMs) with third-party apps that are invoked by a system LLM using interleaved planning and execution phases to answer user queries. These systems introduce new attack vectors where malicious apps can cause integrity violation of planning or execution, availability breakdown, or privacy compromise during execution. In this work, we identify new attacks impacting the integrity of planning, as well as the integrity and availability of execution in LLM-integrated apps, and demonstrate them against IsolateGPT, a recent solution designed to mitigate attacks from malicious apps. We propose Abstract-Concrete-Execute (ACE), a new secure architecture for LLM-integrated app systems that provides security guarantees for system planning and execution. Specifically, ACE decouples planning into two phases by first creating an abstract execution plan using only trusted information, and then mapping the abstract plan to a concrete plan using installed system apps. We verify that the plans generated by our system satisfy user-specified secure information flow constraints via static analysis on the structured plan output. During execution, ACE enforces data and capability barriers between apps, and ensures that the execution is conducted according to the trusted abstract plan. We show experimentally that ACE is secure against attacks from the InjecAgent and Agent Security Bench benchmarks for indirect prompt injection, and our newly introduced attacks. We also evaluate the utility of ACE in realistic environments, using the Tool Usage suite from the LangChain benchmark. Our architecture represents a significant advancement towards hardening LLM-based systems using system security principles.